Error using FITPARAMS

[BorgChick]

Hi,

I am trying to fit my transformation coefficients using fitparams and following the steps outlined in Massey & Davis 1992, "A Users Guide to Stellar CCD Photometry with IRAF." When I get to the step using FITPARAMS, I get the following error:

"ERROR: bus error"

I tried increasing my stacksize by entering

"limit stacksize unlimited"

since we are using tcsh. Is this right?

Do you have any thoughts/ideas why this might be occurring? Thanks!

[fitz]

The stacksize limit applies only to linux systems, and if it were the problem, would affect all tasks (e.g. IMSTAT) equally.

Otherwise, there are no open issues with the task and there's not enough information to comment specifically: a 'bus error' is sometimes a coding problem, but can also be caused by particular parameter setting or data values. What IRAF version and architecture is this? Can you upload sample data and parameters you used to the anonftp at ftp://iraf.noao.edu/pub so we can try to reproduce the problem?

[BorgChick]

We are using a Mac OSX 10.6.8 with verson 2.15.1.a of IRAF. We will post things to the ftp site soon...

Just a quick note: We are using PG0231 as our standard field, and therefore have 7 stars that we are using instead of 1, which is what the guide uses. We weren't sure if our work-around was causing a problem. Also, we used IDL to create a file that looked like Fig. 20 (file called 'standobs') because we did photometry in a different way than that described in the guide. Therefore, we can only really upload the files we used, not the actual data. However, we could potentially upload the individual fits images if you require them.

[BorgChick]

We have uploaded our files to the /pub directory. They are as follows:

fitparam.parameters = parameters used for FITPARAMS
standobs = lists field by star, filter, airmass, xcent, ycent, mag, merr
standstar = lists matched observations of stars
fstandobs.dat = format of standobs file (made to look like fig. 21 in the guide)
testnuro.cfg = made by running mkconfig

[fitz]

The good news is that with your data I was able to reproduce the problem under v2.15 and traced the problem to a pointer not being properly cast for 64-bit architectures. I fixed this and similar instances however there are some pretty strange data structures in PHOTCAL and I suspect there may be additional cases that still need fixing. Anyway, the patched binary can be found in

ftp://iraf.noao.edu/iraf/v215/support/macintel/x_photcal.e

You should download this and install in your $iraf/noao/bin.macintel directory.

The bad news, however, is that this doesn't fix the problem. With the patched binary (and under v216, and under v215 with 32-bit binaries), I get the message "ERROR: No observations could be matched with the catalog entries". The testnuro.ans file contains several status codes indicating "Too few points" and lots of zero values that you should probably look at.

I'm not too familiar with the package but I hope this gives you something to work with.

[BorgChick]

Thanks! We will definitely look into this further and let you know what we find out. Thanks so much for your help!

[BorgChick]

Hi! We were able to install the new binary, and discovered that we had a formatting error in our 'standobs' file. Once that was fixed, we also realized that the catalog name should have been 'onlandolt'. We ran FITPARAMS and got a reasonable output and when we hit 'q' to move onto the next filter, we got a 'BUS ERROR' message again. Any thoughts? We will upload a new copy of our standobs file. Thanks!

[fitz]

Found the problem and updated the binary at

ftp://iraf.noao.edu/iraf/v215/support/macintel/x_photcal.e

With this I was able to get the task to exit normally. BTW, even without the fix the task ran normally under v2.16 but will be fixed in the next release as well.